An experimental study of the effect of feedback from pupils on the behavior of teachers

Abstract

The main purpose of this study was to find out the effect of feedback from pupils on the behavior of teachers. For a statistical approach to the problem, the following research hypotheses were advanced: 1. Teachers receiving feedback from pupils would show more desirable change in behavior than teachers not receiving feedback. 2. More teachers would react positively than negatively to feedback from pupils. 3. Sex, educational training, age, teaching experience, and grade taught would be associated with the amount of change in teachers’ behavior after the feedback. 4. Positive change would result after the feedback in each of the four dimensions of teacher merit. The study was conducted from July to December, 1971, in all public elementary schools of the districts of Bugasong, Patnoñgon, and Sibalom North in the Division of Antique. The subjects were Grades IV, V, and VI teachers who composed the experimental and the control groups. Each group had forty-five teachers and each teacher had an average of twenty-one pupils in his class. During the pretest, the pupils evaluated the behavior of their respective teachers on the Hiligaynon rating instrument consisting of forty items grouped into four behavior dimensions: affective, cognitive, disciplinary, and motivational-innovative merits. The instrument was validated by competent educators and found to have test reliability coefficient which was adequate for the purpose of this study. After the pretest, only the teachers in the experimental group were informed of the ratings given them by their own pupils. This was the feedback treatment which was in the form of a graphic rating scale. Twelve weeks after the feedback, a posttest on the rating instrument was administered to the pupils who had participated in the pretest. Arbitrary values were assigned to the categories of response to compute for the scores of teachers. Each research hypothesis was tested against the null hypothesis at the .05 level of significance. The following statistical procedures were used: 1. Test for the significance of the difference between means and between standard deviations to compare the pretest and the posttest scores within a group under the one-tailed test, and to compare the differential scores of two different groups under the two-tailed test. 2. Test for the significance of the difference between percentages of teachers with increased ratings and with decreased ratings in the experimental group. 3. Test for the significance of the difference between any pair of means by analysis of variance on the posttest scores of the three categorized groups of teaching experience and of grade taught. The findings were as follows: 1. The differential scores of the experimental group showed a significantly higher mean and greater standard deviation than those of the control group. 2. There was a significantly higher percentage of teachers with increased ratings than those with decreased ratings in the experimental group. 3. The differential scores of the male and the female teachers did not show any significant differences. 4. The differential scores of the BSEED and the ETC teachers did not show any significant difference between their means, although the standard deviation of the BSEED group was significantly greater than that of the ETC group. 5. The differential scores of the young teachers showed a significantly higher mean than those of the old teachers, although there was no significant difference between their standard deviations. 6. There was no significant difference between any pair of means shown by analysis of variance among the three groups categorized by the number of years of teaching experience and among those by grade taught. 7. There was a significant increase in the mean, although there was no significant change in the standard deviation, from the pretest to the posttest, in the experimental group as a whole, and within each behavior dimension of the experimental group. 8. The pretest and the posttest scores of the control group did not show any significant difference between their means, although the standard deviation of the pretest was significantly greater than that of the posttest. The foregoing findings led to the following conclusions: 1. Feedback from pupils positively changed the behavior of teachers. 2. Significantly more teachers reacted positively than negatively to feedback from pupils. 3. Teachers who were given feedback—whatever their sex, educational training, teaching experience or teaching xvii responsibility—were significantly affected by evaluation. 4. The young teachers were significantly more receptive to feedback from pupils than were the old teachers. 5. Feedback from pupils produced desirable change in all dimensions of teacher merit; namely, affective, cognitive, disciplinary, and motivational-innovative merits. Based on the findings, the following are recommended: 1. Pupil evaluation of teacher behavior should be used by teachers as a means for self-improvement. 2. The use of feedback should be extended to different groups of workers. 3. A rating instrument similar to the one used in this study should be developed and tested under different conditions to aim for higher reliability and validity of the measuring instrument. 4. More studies on feedback should be conducted on larger populations by other researchers so as to convince school authorities that this type of evaluation is not only necessary but beneficial.